We have compelling new data demonstrating that providing the influenza vaccine in different contexts can induce dramatic shifts in long-term B cell memory to influenza. This important proof of concept suggests that human B cell responses can be re-trained as needed to improve vaccine efficacy. In this project we will explore the mechanism of these observations. The principles gleaned will be important in the development of an improved influenza vaccine. First, in Aim 1, we will explore the role of B cell memory and immune sera in redirecting antibody responses to particular epitopes upon re-exposure to variant influenza strains. Secondly, in Aim 2 we will determine if, like the response to hemagglutinin, a predominance of broadly cross-reactive B cells were activated against neuraminidase upon first exposure to the 2009 pandemic strain of influenza. We will also use novel approaches to determine how stable previously observed biased influenza epitope targeting in human populations is. Finally, we have identified what we believe is an epidemic-memory B cell differentiation that we hypothesize persists at high numbers for months rather than permanently, and provides immediate protection due to the highly likely re-exposure that will occur during an epidemic. We believe that these memory cells, when induced by vaccination with relevant or irrelevant strains, can substantially impact responses to variant influenza strains occurring in the near future in an original-antigenic sin (OAS) fashion. In fact, these cells may account for the dramatic shifts that we observed for some but not all members of our cohort, leading to responses with either greater breadth or more strain-specific responses. We will explore the phenotype and function of these cells. Further, by determining the specificity of the B cells from this population we will directly determine the impact of these memory cells upon re-exposure to a divergent influenza strain. Understanding the types and stages of B cell memory will allow better targeting of the influenza vaccine response. Finally, in the context of these studies we will develop new technologies such as our epitope-shadowing approach using our novel mAb microarray platform to allow rapid and global screening of both monoclonal and polyclonal (serum) epitope targeting. We will also generate a large panel of human mAbs against influenza. These antibodies may identify epitopes important for developing a broadly protective vaccine or the mAbs themselves may have therapeutic potential.